A method and system for scheduling satellite-to-ground laser links
By constructing a two-layer robust model and transforming it into a single-layer mixed integer linear programming, the uncertainty problem in the scheduling of satellite-to-ground laser links is solved, achieving efficient scheduling under uncertain environments and ensuring system throughput and resource utilization.
Patent Information
- Application Number
- CN202611614099.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-10-19
- Publication Date
- 2026-11-17
AI Technical Summary
Existing satellite-to-ground laser link scheduling methods fail to effectively handle the uncertainty in link availability and data transmission rate caused by cloud cover and atmospheric turbulence, resulting in system throughput loss.
A two-layer robust model is constructed and transformed into a single-layer mixed-integer linear programming model through linearization and KKT conditions. This model independently represents link availability and data transmission rate deviation. The product terms are linearized using the McCormick envelope, and link scheduling is optimized to maximize system throughput.
In uncertain environments, robustness of system throughput is guaranteed, avoiding overestimation of scheduling schemes due to neglecting coupling effects, and improving resource utilization and disturbance resistance.
Smart Images

Figure CN123122659A_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the field of satellite-to-ground laser communication network resource scheduling technology, and particularly relates to a satellite-to-ground laser link scheduling method and system. Background Technology
[0002] With the continuous development of applications such as satellite remote sensing, space science exploration, satellite internet, and space-ground integrated networks, the scale of observation data, exploration data, and operational data generated by satellites in orbit continues to expand, placing higher demands on high-speed data backhaul capabilities between satellites and ground stations. Space-to-ground laser communication has advantages such as high transmission rates, abundant spectrum resources, and small beam divergence angles, providing an effective transmission means for high-speed space-to-ground communication networks composed of multiple satellites, multiple ground stations, and multiple laser communication terminals. Given that the locations and configurations of ground stations are already determined, unified scheduling of candidate laser links between satellites and ground stations within each scheduling time slot is required. The link scheduling scheme directly determines the system throughput throughout the entire scheduling cycle.
[0003] Existing satellite-to-ground laser link scheduling methods typically use predicted link availability and predicted data transmission rate as accurately known fixed parameters. A deterministic optimization model is constructed based on the nominal predicted values, aiming to maximize system throughput. Solving this model determines the actual satellite-to-ground laser links established within each scheduling time slot. However, the performance of satellite-to-ground laser communication is simultaneously affected by cloud cover and atmospheric turbulence. Cloud cover directly leads to link interruptions, affecting link availability. When a link is available, atmospheric turbulence causes intensity flicker and wavefront distortion at the receiver, degrading signal quality and reducing data transmission rate. Due to the randomness of cloud movement, dissipation, and turbulence intensity changes, there may be discrepancies between the predicted values and the actual values during the scheduling execution phase. If the scheduling scheme selects links solely based on nominal values, it may preferentially allocate a limited number of laser communication terminals to links with lower actual availability or lower actual transmission rates, resulting in a loss of system throughput. Summary of the Invention
[0004] This application provides a satellite-to-ground laser link scheduling method and system. By constructing a two-layer robust model that selects link establishment variables in the outer layer and selects link availability deviation and data transmission rate deviation in the inner layer, and transforming it into a single-layer mixed integer linear programming model for solution by means of linearization, KKT conditions and large constant method, this method solves the problems in the prior art that cannot simultaneously characterize the uncertainty of link availability and data transmission rate, have difficulty in handling the non-convex bilinear terms caused by the coupling of the two types of deviations, and have difficulty in directly and efficiently solving the model due to the conflict between inner and outer target layers.
[0005] Firstly, this application provides a satellite-to-ground laser link scheduling method, including: The effective visibility duration, nominal link availability, nominal data transmission rate, maximum allowable deviation of link availability, maximum allowable deviation of data transmission rate, first robust control parameter, second robust control parameter, upper limit of the number of satellite-to-ground laser links that each satellite can establish simultaneously in any time slot, and upper limit of the number of satellite-to-ground laser links that each ground station can access simultaneously in any time slot are obtained in the satellite-to-ground laser communication network. The system throughput and scheduling constraints are constructed using binary link establishment variables. The system throughput is determined based on the effective visibility duration, actual link availability, and actual data transmission rate. The scheduling constraints include the upper limit constraint on the number of satellite-to-ground laser links that each satellite can establish simultaneously in any time slot, the upper limit constraint on the number of satellite-to-ground laser links that each ground station can access simultaneously in any time slot, and the value constraints of the binary link establishment variables. The actual link availability and actual data transmission rate are expressed as the sum of nominal link availability and the deviation of link availability, and the sum of nominal data transmission rate and the deviation of data transmission rate, respectively. Based on nominal link availability, nominal data transmission rate, maximum allowable deviation of link availability, maximum allowable deviation of data transmission rate, first robust control parameter and second robust control parameter, a set of uncertain budgets for link availability and data transmission rate is constructed. A two-layer robust model is constructed using system throughput, link availability budget uncertainty set, and data transmission rate budget uncertainty set as the outer layer of binary link establishment variables and the inner layer of link availability deviation and data transmission rate deviation. To address the inner-layer optimization problem in the two-layer robust model, the link availability deviation and data transmission rate deviation are decomposed into positive and negative deviation variables respectively to linearize the absolute value constraint. Bilinear auxiliary variables are introduced and the product of link availability deviation and data transmission rate deviation is linearized using the McCormick envelope. Lagrangian functions and KKT conditions are constructed for the linearized inner-layer optimization problem. The complementary relaxation conditions are linearized using the large constant method. The linearized inner-layer conditions and scheduling constraints are then merged into a single-layer mixed-integer linear programming model. Solve the single-layer mixed integer linear programming model and output the satellite-to-ground laser link scheduling results.
[0006] Optionally, system throughput and scheduling constraints are constructed using binary linked variables, including: Construct the expression for system throughput and scheduling constraints: ; Where ST is the system throughput; I is the total number of scheduling time slots; M is the total number of satellites in the space-to-ground laser communication network; K is the total number of ground stations in the space-to-ground laser communication network; Δt i,m,kφ represents the effective visible duration of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; i = 1, 2, ..., I; m = 1, 2, ..., M; k = 1, 2, ..., K; φ i,m,k η represents the actual data transmission rate of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; i,m,k x represents the actual link availability of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; i,m,k This represents the on / off status of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; x i,m,k =1 indicates that the link is connected; x i,m,k =0 indicates that the link is disconnected; This means that the number of links established simultaneously by each satellite in any time slot does not exceed the number P; This indicates that the number of links that each ground station can simultaneously access in any time slot does not exceed the number Q.
[0007] Optionally, the actual link availability and actual data transmission rate are expressed as the sum of nominal link availability and the deviation from link availability, and the sum of nominal data transmission rate and the deviation from data transmission rate, respectively. Based on nominal link availability, nominal data transmission rate, maximum permissible deviation of link availability, maximum permissible deviation of data transmission rate, first robust control parameter, and second robust control parameter, a set of uncertainties in the budget for link availability and a set of uncertainties in the budget for data transmission rate are constructed, including: Based on the maximum permissible deviation 'a' of the link availability of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot. i,m,k Construct a link availability budget uncertainty set U1 with the first robust control parameter Γ1; where the link availability budget uncertainty set U1 contains |δ i,m,k |≤a i,m,k , 0≤η i,m,k ≤1, and ;δ i,m,k η represents the link availability deviation of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; i,m,k The actual link availability of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; I is the total number of scheduling time slots; M is the total number of satellites in the satellite-to-ground laser communication network; K is the total number of ground stations in the satellite-to-ground laser communication network; i=1,2,…,I; m=1,2,…,M; k=1,2,…,K; |·| represents taking the absolute value; Based on the maximum permissible deviation r of the data transmission rate of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot. i,m,k The second robust control parameter Γ2 is used to construct the data transmission rate budget uncertainty set U2; where the data transmission rate budget uncertainty set U2 includes: |ζ i,m,k |≤r i,m,k φ i,m,k ≥0, and ;ζ i,m,k φ represents the signed data transmission rate deviation of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; i,m,k The actual data transmission rate of the candidate satellite-to-ground laser link consisting of the m-th satellite and the k-th ground station within the i-th scheduling time slot is given.
[0008] Optionally, a two-layer robust model is constructed using system throughput, the set of uncertainties regarding link availability, and the set of uncertainties regarding data transmission rate, comprising an outer layer of binary link establishment variables for selection, and an inner layer of variable variables for selection regarding link availability and data transmission rate. This model includes: The expression for constructing the two-layer robust model f is: ; Where, x i,m,k This represents the on / off status of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; x i,m,k =1 indicates that the link is connected; x i,m,k =0 indicates that the link is disconnected; This means that the number of links established simultaneously by each satellite in any time slot does not exceed the number P; This indicates that the number of links simultaneously accessed by each ground station in any time slot does not exceed the number Q; δ i,m,k ζ represents the link availability deviation of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; i,m,k Δt represents the signed data transmission rate deviation of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot. i,m,k The effective visibility duration of the candidate satellite-to-ground laser link consisting of the m-th satellite and the k-th ground station within the i-th scheduling time slot; The nominal link availability is the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot. Let be the nominal data transmission rate of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; i = 1, 2, ..., I; m = 1, 2, ..., M; k = 1, 2, ..., K; the link availability budget uncertainty set U1 contains |δ i,m,k|≤a i,m,k , 0≤η i,m,k ≤1, and Γ1 is the first robust control parameter; a i,m,k Let |ζ| represent the maximum permissible deviation in the link availability of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; the data transmission rate budget uncertainty set U2 includes: |ζ| i,m,k |≤r i,m,k φ i,m,k ≥0, and ;r i,m,k Γ2 represents the maximum permissible deviation of the data transmission rate of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; Γ2 is the second robust control parameter. This means that when δ is satisfied i,m,k and ζ i,m,k Find the minimum value of the expression within the parentheses among all possible values; This means that when x is satisfied i,m,k Find the maximum value of the expression within the parentheses from all possible values.
[0009] Optionally, for the inner-layer optimization problem in the two-layer robust model, the link availability deviation and data transmission rate deviation are decomposed into positive and negative deviation variables respectively to linearize the absolute value constraint. A bilinear auxiliary variable is introduced, and the product of the link availability deviation and data transmission rate deviation is linearized using the McCormick envelope, including: For each candidate link, a non-negative link availability positive deviation variable is introduced. and negative deviation variables and non-negative data transmission rate positive deviation variable and negative deviation variables ;in, ; ; ; ; Using the McCormick envelope, according to The range of values is [-a] i,m,k ,a i,m,k ]and The range of values is [-r i,m,k ,r i,m,k Constructing linear constraints: ; in, ;ψ i,m,k Let be the bilinear auxiliary variable corresponding to the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot.
[0010] Optionally, a Lagrangian function and KKT conditions are constructed for the linearized inner-layer optimization problem, the complementary relaxation conditions are linearized using the large constant method, and the linearized inner-layer conditions and scheduling constraints are merged into a single-layer mixed-integer linear programming model, including: The linearized inner-layer optimization problem is expressed as follows: the link availability positive and negative deviation variables, the data transmission rate positive and negative deviation variables, and the bilinear auxiliary variable are used as the inner-layer original variables. The goal is to minimize the system throughput, and the conditions are met by the deviation boundary constraints, two types of robust budget constraints, actual link availability value constraints, actual data transmission rate non-negativity constraints, and McCormick envelope constraints. For each constraint of the inner-level optimization problem, a non-negative dual variable is introduced to construct a Lagrangian function; the Lagrangian function consists of the sum of the objective function of the inner-level optimization problem and the summation of each constraint multiplied by its corresponding dual variable. The KKT conditions are constructed based on the Lagrangian function. The KKT conditions include: primal feasibility conditions, which are that all constraints of the inner optimization problem are satisfied; dual feasibility conditions, which are that each dual variable is non-negative; stationary point conditions, which are that the partial derivatives of the Lagrangian function with respect to the inner primal variables are set to zero; and complementary relaxation conditions, which are that the product of each dual variable and the corresponding constraint relaxation is zero, and the constraint relaxation is expressed by the inner primal variables. For each complementary relaxation condition, an auxiliary binary variable is introduced, and the large constant method is used to linearize the complementary relaxation condition into two linear inequalities; the large constant is a positive constant greater than the possible values of the dual variable and the constraint relaxation amount. By merging scheduling constraints, primal feasible conditions, dual feasible conditions, stationary conditions, and linearized complementary relaxation conditions, a single-level mixed-integer linear programming model is formed. The single-level mixed-integer linear programming model uses binary link-building variables, inner-level primal variables, dual variables, and auxiliary binary variables as decision variables, with the objective of maximizing the system throughput under the worst-case link state. The worst-case link state is the value of link availability deviation and data transmission rate deviation that minimizes the system throughput within the two sets of budget uncertainty.
[0011] Optionally, a single-layer mixed-integer linear programming model is solved and the satellite-to-ground laser link scheduling results are output, including: The single-layer mixed-integer linear programming model is input into the mixed-integer linear programming solver for solving, and the values of the binary chain-building variables are obtained. The link establishment target for each scheduling time slot is determined based on the value of the binary link establishment variable. When the binary link establishment variable corresponding to the candidate satellite-to-ground laser link composed of the m-th satellite and the k-th ground station in the i-th scheduling time slot is 1, the candidate satellite-to-ground laser link between the m-th satellite and the k-th ground station is taken as the link establishment target for the i-th scheduling time slot. When the binary link establishment variable is 0, the corresponding candidate satellite-to-ground laser link is not taken as the link establishment target. Output the list of link establishment objects for each scheduling time slot in the order of scheduling time slots, as the result of satellite-to-ground laser link scheduling.
[0012] Secondly, this application provides a satellite-to-ground laser link scheduling system, comprising: The acquisition module is used to acquire the effective visibility duration, nominal link availability, nominal data transmission rate, maximum allowable deviation of link availability, maximum allowable deviation of data transmission rate, first robust control parameter, second robust control parameter, upper limit of the number of satellite-to-ground laser links that each satellite can establish simultaneously in any time slot, and upper limit of the number of satellite-to-ground laser links that each ground station can access simultaneously in any time slot in the satellite-to-ground laser communication network. The first construction module is used to construct the system throughput and scheduling constraints represented by binary link establishment variables. The system throughput is determined based on the effective visibility duration, actual link availability, and actual data transmission rate. The scheduling constraints include the upper limit constraint on the number of satellite-to-ground laser links that each satellite can establish simultaneously in any time slot, the upper limit constraint on the number of satellite-to-ground laser links that each ground station can access simultaneously in any time slot, and the value constraints of the binary link establishment variables. The second construction module is used to express the actual link availability and the actual data transmission rate as the sum of the nominal link availability and the deviation of the link availability, and the sum of the nominal data transmission rate and the deviation of the data transmission rate, respectively. Based on the nominal link availability, the nominal data transmission rate, the maximum allowable deviation of the link availability, the maximum allowable deviation of the data transmission rate, the first robust control parameter, and the second robust control parameter, the link availability budget uncertainty set and the data transmission rate budget uncertainty set are constructed. The third building module is used to construct a two-layer robust model with outer layer selection binary link establishment variables and inner layer selection link availability deviation and data transmission rate deviation, based on system throughput, link availability budget uncertainty set and data transmission rate budget uncertainty set; The linearization module is used for inner-layer optimization problems in a two-layer robust model. It decomposes the link availability deviation and data transmission rate deviation into positive and negative deviation variables respectively to linearize the absolute value constraints. It introduces bilinear auxiliary variables and uses the McCormick envelope to linearize the product of the link availability deviation and the data transmission rate deviation. The fourth module is used to construct the Lagrangian function and KKT conditions for the linearized inner-layer optimization problem, linearize the complementary relaxation conditions using the large constant method, and merge the linearized inner-layer conditions and scheduling constraints into a single-layer mixed-integer linear programming model. The scheduling module is used to solve a single-layer mixed integer linear programming model and output the scheduling results of the satellite-to-ground laser link.
[0013] Thirdly, this application provides a computer device including a processor and a memory; wherein, when the processor executes a computer program stored in the memory, it implements the steps of the satellite-to-ground laser link scheduling method described in the first aspect.
[0014] Fourthly, this application provides a computer-readable storage medium for storing a computer program; when the computer program is executed by a processor, it implements the steps of the satellite-to-ground laser link scheduling method described in the first aspect.
[0015] Beneficial effects: This application represents actual link availability as the sum of nominal link availability and link availability deviation, and actual data transmission rate as the sum of nominal data transmission rate and data transmission rate deviation. It also constructs separate sets for link availability budget uncertainty and data transmission rate budget uncertainty, thus achieving independent characterization and joint optimization of the two types of uncertainty in the model structure. Since link throughput is determined by the product of effective visibility duration, link availability, and data transmission rate, the link availability deviation and data transmission rate deviation are coupled to the system throughput through the product term. This application introduces a bilinear auxiliary variable and uses the McCormick envelope to linearize this product term, fully preserving the coupling effect of the two types of deviations in the worst-case perturbation solution process. This allows the outer scheduling decision to fully perceive the joint impact of multi-source deviations, avoiding overestimation of link transmission capacity by the scheduling scheme due to ignoring the coupling effect.
[0016] In the two-layer robust model constructed in this application, the inner layer optimization searches for the combination of deviations that minimizes system throughput within two sets of budget uncertainties, i.e., constructing the worst-case link state; the outer layer optimization selects the link establishment combination that maximizes system throughput under the constraint of the number of concurrent links at the satellite and ground stations. This maximization-minimization optimization structure ensures that the obtained scheduling scheme can still achieve a high lower limit of throughput under the most unfavorable combination of deviations, which is equivalent to providing a robust guarantee for system throughput. This guarantee is not based on empirical safety margin settings, but is automatically obtained through the worst-case search mechanism inherent in the optimization model. Therefore, it can quantitatively characterize the disturbance resistance of the scheduling scheme without relying on manually setting conservative thresholds.
[0017] The first and second robust control parameters control the overall disturbance budget for the uncertain sets of link availability and data transmission rate, respectively. Their values determine the upper limit of the number of links allowed to deviate and the upper limit of the deviation magnitude. When the robust control parameter is zero, both types of deviations are forced to zero, and the model degenerates into a deterministic scheduling model based on nominal values. In this case, the scheduling scheme has no robustness constraints, has the highest resource utilization, but lacks disturbance resistance. When the robust control parameter is at its maximum value, the deviation of each link can freely take values within its maximum allowable deviation range. The model has the strongest disturbance resistance but may sacrifice some throughput. By independently adjusting the two parameters, differentiated configurations can be made between link availability robustness and data transmission rate robustness based on task reliability requirements, link prediction accuracy, and network resource conditions. For example, for links in areas with frequent cloud changes, the first robust control parameter can be increased to enhance the ability to withstand availability fluctuations; for links with high turbulence prediction accuracy, the second robust control parameter can be decreased to reduce unnecessary conservatism. This fine-grained conservatism adjustment mechanism overcomes the shortcomings of insufficient conservatism control granularity in traditional single-parameter robust models, enabling scheduling schemes to better adapt to the differentiated uncertainty characteristics of different links and time slots.
[0018] This application first decomposes the link availability deviation and data transmission rate deviation into positive and negative deviation variables, respectively, and uses a linear combination of non-negative variables to replace the absolute value constraints. Then, it introduces bilinear auxiliary variables and uses the McCormick envelope to construct linear constraints, transforming the deviation product term into a linear expression. Next, it constructs Lagrangian functions and KKT conditions for the inner-layer linearized optimization problem, transforming the inner-layer optimality conditions into constraints for the outer-layer model. Finally, it uses the large constant method to linearize the complementary relaxation conditions. After the above processing, the original two-layer robust model is completely transformed into a single-layer mixed-integer linear programming model. This model can be directly solved globally using mature commercial solvers without the need to design specialized decomposition or heuristic algorithms, thus significantly reducing the engineering implementation difficulty while ensuring solution accuracy.
[0019] The budget uncertainty set employed in this application allows only a limited number of links to deviate, and the deviation magnitude does not exceed the maximum allowable deviation magnitude for each link. This budget mechanism restricts the search range for the worst-case scenario to a reasonable range, avoiding excessive influence of low-probability extreme events on scheduling decisions. Furthermore, since the model optimizes system throughput rather than deviation minimization, the scheduling scheme strives to maximize total data transmission while mitigating uncertainty, thus achieving a better balance between robustness and resource utilization. Attached Figure Description
[0020] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0021] Figure 1 A schematic flowchart illustrating a satellite-to-ground laser link scheduling method provided in this application embodiment; Figure 2 A comparison chart of system throughput under different scheduling methods provided in the embodiments of this application; Figure 3 This is a schematic diagram of a satellite-to-ground laser link scheduling system provided in an embodiment of this application. Detailed Implementation
[0022] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.
[0023] This application addresses a satellite-to-ground laser communication network consisting of multiple satellites, multiple ground stations, and a ground control center. Both the satellite and ground station ends are equipped with a limited number of laser communication terminals. The scheduling cycle is divided into several time slots, within which several candidate satellite-to-ground laser links are formed between the satellites and ground stations. Due to cloud cover and atmospheric turbulence, both the availability of candidate links and data transmission rates exhibit prediction biases. This application incorporates both types of biases into the scheduling model, describes the bias range using a budget uncertainty set, constructs a two-layer robust model (outer layer selects links, inner layer selects the worst-case bias), and then transforms it into a single-layer mixed-integer linear programming model for solution using linearization, KKT conditions, and the large constant method.
[0024] Example 1 like Figure 1 As shown, this application provides a satellite-to-ground laser link scheduling method, including: Step 101: Obtain the effective visibility duration, nominal link availability, nominal data transmission rate, maximum allowable deviation of link availability, maximum allowable deviation of data transmission rate, first robust control parameter, second robust control parameter, upper limit of the number of satellite-to-ground laser links that each satellite can establish simultaneously in any time slot, and upper limit of the number of satellite-to-ground laser links that each ground station can access simultaneously in any time slot in the satellite-to-ground laser communication network.
[0025] In this embodiment, the satellite-to-ground laser communication network includes M satellites and K ground stations. The scheduling cycle is divided into I scheduling time slots. A predetermined number of laser communication terminals are configured at both the satellite end and the ground station end.
[0026] Candidate satellite-to-ground laser links are uniquely identified by a triple (i,m,k), which is the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station in the i-th scheduling time slot, or the candidate satellite-to-ground laser link between the m-th satellite and the k-th ground station in the i-th scheduling time slot, where i=1,2,…,I; m=1,2,…,M; k=1,2,…,K.
[0027] Effective visible duration Δt i,m,k It can be obtained by orbit simulation software (such as STK), orbit extrapolation, or satellite-to-ground station visibility window calculation, representing the length of time (in seconds) during which the candidate link (i,m,k) is visible and available in the i-th scheduling time slot; the actual data transmission rate φ is calculated. i,m,k η represents the data transmission rate of candidate link (i,m,k) in an available state, in megabits per second (Mb / s) or gigabits per second (Gb / s); actual link availability η i,m,k The variable x represents the proportion or availability of candidate link (i,m,k) in the i-th scheduling time slot, is dimensionless, and ranges from [0,1]; the binary link establishment variable x i,m,k Indicates whether the candidate link (i,m,k) is established, x i,m,k =1 indicates establishment, x i,m,k =0 indicates that it is not created.
[0028] Nominal link availability (i.e., the nominal link availability of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot, which can be obtained from cloud cover statistics, short-term cloud image prediction, or numerical weather prediction) and nominal data transmission rate. (i.e., the nominal data transmission rate of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot, which can be obtained from the link budget, atmospheric turbulence state prediction, or historical measured rate) represent the predicted link availability and data transmission rate, respectively; the link availability deviation δ i,m,k Represents the actual link availability minus the nominal link availability, dimensionless; data transmission rate deviation ζ i,m,k This represents the actual data transmission rate minus the nominal data transmission rate, with the same dimensions as the data transmission rate; the maximum permissible deviation 'a' for link availability. i,m,k (This can be determined based on historical prediction error quantiles, upper limits of engineering experience, or prediction reliability for different sites and time slots; for example, taking a...) i,m,k =0.1) Dimensionless; Maximum permissible deviation of data transmission rate ri,m,k (This can be determined based on the historical rate fluctuation quantiles, for example, taking r...) i,m,k =10Mb / s) has the same dimensions as the data transmission rate; the first robust control parameter Γ1 and the second robust control parameter Γ2 can both be set according to the mission reliability requirements, with values ranging from 0≤Γ1≤I·M·K and 0≤Γ2≤I·M·K respectively. Γ1 and Γ2 are dimensionless and control the overall disturbance budget of the two types of uncertain sets respectively; P represents the upper limit of the number of satellite-to-ground laser links that each satellite can establish simultaneously in any time slot; Q represents the upper limit of the number of satellite-to-ground laser links that each ground station can access simultaneously in any time slot.
[0029] Step 102: Construct system throughput and scheduling constraints represented by binary link establishment variables; system throughput is determined based on effective visibility duration, actual link availability, and actual data transmission rate; scheduling constraints include the upper limit constraint on the number of satellite-to-ground laser links that each satellite can establish simultaneously in any time slot, the upper limit constraint on the number of satellite-to-ground laser links that each ground station can access simultaneously in any time slot, and the value constraints of the binary link establishment variables.
[0030] In this embodiment, the system throughput ST represents the total amount of data transmitted from all scheduled satellites to the ground station within the scheduling period, in megabits (Mb) or gigabits (Gb).
[0031] For example, construct an expression for system throughput: .
[0032] Where I is the total number of scheduling time slots; M is the total number of satellites in the space-to-ground laser communication network; K is the total number of ground stations in the space-to-ground laser communication network; Δt i,m,k φ represents the effective visibility duration of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; i,m,k η represents the actual data transmission rate of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; i,m,k x represents the actual link availability of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; i,m,k This represents the on / off status of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; x i,m,k =1 indicates that the link is connected; x i,m,k =0 indicates that the link is disconnected.
[0033] Construct scheduling constraints: This means that the number of links established simultaneously by each satellite in any time slot does not exceed the number P; This indicates that the number of links that each ground station can simultaneously access in any time slot does not exceed the number Q.
[0034] Step 103: Express the actual link availability and actual data transmission rate as the sum of nominal link availability and link availability deviation, and the sum of nominal data transmission rate and data transmission rate deviation, respectively. Construct the link availability budget uncertainty set and the data transmission rate budget uncertainty set based on nominal link availability, nominal data transmission rate, maximum allowable deviation of link availability, maximum allowable deviation of data transmission rate, first robust control parameter and second robust control parameter.
[0035] In this embodiment, the actual link availability is expressed as the sum of the deviations between the nominal link availability and the signed link availability: ;δ i,m,k It represents the link availability deviation of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot.
[0036] Based on the maximum permissible deviation 'a' of the link availability of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot. i,m,k Construct a link availability budget uncertainty set U1 with the first robust control parameter Γ1; where the link availability budget uncertainty set U1 contains |δ i,m,k |≤a i,m,k , 0≤η i,m,k ≤1, and ; |·| represents taking the absolute value; if a i,m,k =0, then δ i,m,k =0, corresponding to the normalized budget item Process as zero.
[0037] The actual data transfer rate is expressed as the sum of the deviations between the nominal data transfer rate and the signed data transfer rate: ;ζ i,m,k It represents the signed data transmission rate deviation of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot.
[0038] Based on the maximum permissible deviation r of the data transmission rate of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot. i,m,k The second robust control parameter Γ2 is used to construct the data transmission rate budget uncertainty set U2; where the data transmission rate budget uncertainty set U2 includes: |ζ i,m,k |≤r i,m,k φ i,m,k ≥0, and If r i,m,k =0, then ζ i,m,k =0, corresponding to the normalized budget item Treat it as zero; when Γ1=Γ2=0, both types of deviations are zero, and the model degenerates into a deterministic scheduling model based on the nominal link state.
[0039] The first robust control parameter Γ1 controls the overall disturbance budget for the uncertain set of link availability; the second robust control parameter Γ2 controls the overall disturbance budget for the uncertain set of data transmission rate. These two types of parameters are set separately, thus allowing independent adjustment of the model's conservatism for cloud prediction confidence and turbulence prediction confidence. When Γ1=0, all link availability deviations are forced to zero; when Γ1=I·M·K, each link availability deviation can take values within its maximum permissible deviation range. The second robust control parameter Γ2 has the same budget control meaning.
[0040] The maximum permissible deviation is determined based on historical prediction error quantiles, upper limits of engineering experience, or prediction reliability for different sites and time slots. For links with rapid cloud changes or large turbulence prediction errors, a larger maximum permissible deviation is set; for links with stable predictions, a smaller maximum permissible deviation is set.
[0041] Step 104: Construct a two-layer robust model using system throughput, link availability budget uncertainty set, and data transmission rate budget uncertainty set, with outer layer selection of binary link establishment variables and inner layer selection of link availability deviation and data transmission rate deviation.
[0042] To maximize system throughput under the worst-case link states within two sets of budget uncertainties, a two-layer robust optimization model f is constructed, consisting of an outer layer of link scheduling and an inner layer of multi-source uncertain disturbances. .
[0043] in, This means that when δ is satisfied i,m,k and ζ i,m,k Find the minimum value of the expression within the parentheses among all possible values; This means that when x is satisfied i,m,k Find the maximum value of the expression within the parentheses from all possible values.
[0044] The inner optimization uses link availability deviation and data transmission rate deviation as decision variables to construct the worst link state that minimizes system throughput within two sets of budget uncertainties. The outer optimization uses binary link scheduling variables as decision variables to select the link establishment combination that maximizes system throughput under the constraints of the number of terminals at the satellite end and the ground station end.
[0045] Expand the product of the two types of uncertain parameters according to the candidate link index as follows: .
[0046] As can be seen from the above formula, in addition to the nominal term and two first-order deviation terms, the formula also includes a bilinear product term of the link availability deviation and the data transmission rate deviation.
[0047] Step 105: For the inner layer optimization problem in the two-layer robust model, the link availability deviation and data transmission rate deviation are decomposed into positive deviation variables and negative deviation variables respectively to linearize the absolute value constraint. Bilinear auxiliary variables are introduced and the product of link availability deviation and data transmission rate deviation is linearized using the McCormick envelope.
[0048] For the link availability deviation of each candidate link, a non-negative positive deviation variable is introduced. Non-negative negative deviation variables ; This indicates the upward deviation. This represents the absolute magnitude of the downward deviation, and is decomposed as follows: ; .
[0049] For the data transmission rate deviation of each candidate link, a non-negative positive deviation variable is introduced. Non-negative negative deviation variables ; This indicates the upward deviation. This represents the absolute magnitude of the downward deviation, and is decomposed as follows: ; .
[0050] By replacing the absolute values in the two types of budget constraints with the sum of the corresponding positive and negative deviation variables, the absolute value constraints are transformed into linear constraints. ; ; ; .
[0051] For the bilinear product term, a bilinear auxiliary variable ψ is introduced. i,m,k (i.e., the bilinear auxiliary variable corresponding to the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station in the i-th scheduling time slot): .
[0052] because The range of values is [-a] i,m,k ,a i,m,k ], The range of values is [-r i,m,k ,r i,m,k Using the McCormick envelope, four sets of linear constraints are constructed: .
[0053] Step 106: Construct the Lagrangian function and KKT conditions for the linearized inner-layer optimization problem, linearize the complementary relaxation conditions using the large constant method, and merge the linearized inner-layer conditions and scheduling constraints into a single-layer mixed-integer linear programming model.
[0054] To transform the bi-level optimization problem into a single-level problem and solve it, we first introduce dual variables corresponding to the constraints of the linearized inner-level optimization problem. The KKT conditions consist of two core components: stationary conditions and complementary relaxation conditions. The Lagrangian function is composed of the original objective function and the weighted constraint terms, where the weights are the dual variables, also known as Lagrange multipliers. The stationary conditions are obtained by taking the partial derivative of the Lagrangian function with respect to the original variables and setting it to zero, reflecting the extreme point characteristics of the objective function under constraints. The complementary relaxation conditions require that the product of each original constraint and its corresponding dual variable be zero, thus establishing an intrinsic link between the original feasibility and the dual optimality.
[0055] Specifically, the linearized inner-layer optimization problem is: using the original inner-layer variables... , , , and ψ i,m,k The decision variables are defined as follows, with the objective of minimizing system throughput, satisfying the following constraints: deviation boundary constraints, two types of budget constraints, actual link availability constraints, actual rate non-negativity constraints, and four sets of McCormick envelope constraints. A non-negative dual variable is introduced for each constraint, and a Lagrangian function is constructed. The Lagrangian function is equal to the sum of the inner objective function plus each constraint multiplied by its corresponding dual variable. Based on the Lagrangian function, the KKT conditions are established: Original feasible condition: All inner constraints are satisfied.
[0056] Duality is feasible if all dual variables are non-negative.
[0057] Stationary conditions: The Lagrange function is given by... , , , and ψ i,m,k Find the partial derivatives and set them to zero.
[0058] Complementary relaxation condition: The product of each dual variable and its corresponding constraint relaxation is zero, and the constraint relaxation is expressed by the inner primitive variable.
[0059] The following nonnegative dual variables are introduced and defined according to their corresponding original constraints as follows: In the dual variable notation, η and φ are used to distinguish between link availability constraints and data transmission rate constraints; the suffixes dmin and dmax represent the lower bound constraint and upper bound constraint of the deviation, respectively, and the suffixes min and max represent the lower bound constraint and upper bound constraint of the physical value of the actual link state, respectively; the subscripts (i,m,k) still follow the aforementioned candidate link index convention.
[0060] Corresponding to the original constraints .
[0061] Corresponding to the original constraints .
[0062] Corresponding to the original constraints .
[0063] Corresponding to the original constraints .
[0064] μ η,i,m,k ≥0 corresponds to the original constraint .
[0065] ν η,i,m,k ≥0 corresponds to the original constraint .
[0066] μ φ,i,m,k ≥0 corresponds to the original constraint .
[0067] ν φ,i,m,k ≥0 corresponds to the original constraint .
[0068] θ1≥0, corresponding to the first robust budget constraint .
[0069] θ2≥0 corresponds to the second robust budget constraint .
[0070] Corresponding to the original constraints .
[0071] Corresponding to the original constraints .
[0072] Corresponding to the original constraints .
[0073] Let represent the nonnegative dual variable corresponding to the q-th McCormick envelope constraint, q=1,2,3,4; the superscript q only indicates the number of the four constraint groups; this dual variable and the bilinear auxiliary variable ψ (without superscript) i,m,k They have different meanings.
[0074] The constraint functions corresponding to the first and second types of robust budget constraints are respectively expressed as follows: ; .
[0075] The constraint functions corresponding to the four sets of McCormick envelope constraints are expressed as follows: ; ; ; .
[0076] All the aforementioned dual variables together form the set Dual. By multiplying each dual variable by its corresponding primal constraint, the constraint weighting terms in the Lagrangian function are formed. Specifically, these include L1 to L... 17 item: ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; .
[0077] The Lagrangian function L is derived from the objective function of the inner optimization problem and the above L1 to L... 17 The sum of the terms constitutes the following: L = L0 + L1 + L2 + ... + L 17 .
[0078] Where L0 is the objective function of the inner optimization problem, i.e., minimizing the system throughput: .
[0079] For the Lagrange function with respect to , , , and Taking the partial derivatives and setting them all to zero, we obtain the stationary point condition: , , , and .
[0080] In addition to the stationary point condition mentioned above, the KKT conditions also include all primal feasibility conditions of the linearized inner-layer optimization problem, all dual feasibility conditions where all dual variables are non-negative, and complementary relaxation conditions where the product of each dual variable and its corresponding primal constraint relaxation is zero, uniformly expressed as: g j (y)≤0, λ j ≥0, λ j g j (y)=0; j=1,2,…,n; y is the set of inner primitive variables; g j (y) is the left-hand function of the j-th original constraint; λ j Let be the nonnegative dual variable corresponding to the j-th primal constraint; n is the total number of primal constraints. The stationary condition, primal feasibility condition, dual feasibility condition, and complementary relaxation condition together characterize the optimal solution of the inner optimization problem given the outer link scheduling variables.
[0081] Complementary relaxation conditions require that at least one of the dual variables and its corresponding primal constraint relaxations be zero, essentially constituting a nonlinear equation. To transform these conditions into linear constraints that can be handled by a commercial mixed-integer programming solver, an auxiliary binary variable is introduced for each complementary relaxation condition, and a large constant linearization method is used for processing. In this embodiment, the sufficiently large constant in the linearization method is uniformly denoted as B.
[0082] Taking the complementary relaxation condition corresponding to the lower bound constraint of link availability deviation as an example, an auxiliary binary variable is introduced. That is, the complementary relaxed linearized auxiliary binary variable corresponding to the lower bound constraint of link availability deviation. And add the following two linear constraints: , .
[0083] when At that time, by and get Since B is a sufficiently large positive integer, It does not affect the original constraints.
[0084] when hour, With original constraints Together ,and There are no restrictions on the feasibility of the original dual. Therefore, the above two sets of constraints guarantee that at least one of the following holds: "the corresponding dual variable is zero" or "the original constraint is active", thus achieving the equivalent linearization of the complementary relaxation conditions.
[0085] Following the same approach (introducing auxiliary binary variables), the complementary relaxation conditions corresponding to the upper bound constraint of link availability deviation are set as follows: , ; It is also an auxiliary binary variable (at this time, it is the complementary relaxed linearized auxiliary binary variable corresponding to the upper bound constraint of link availability deviation).
[0086] For the complementary relaxation conditions corresponding to the first robust budget constraint: , ; This is also an auxiliary binary variable (at this time, it is the complementary relaxed linearized auxiliary binary variable corresponding to the first robust budget constraint).
[0087] For the complementary relaxation conditions corresponding to the first set of McCormick envelope constraints: , ; This is also an auxiliary binary variable (at this time, it is the complementary relaxed linearized auxiliary binary variable corresponding to the first group of McCormick envelope constraints).
[0088] For the lower bound constraint of data transmission rate deviation, the upper bound constraint of data transmission rate deviation, the non-negativity constraint of positive and negative deviation variables, the second robust budget constraint, the upper and lower bound constraints of actual link availability, the non-negativity constraint of actual data transmission rate, and the complementary relaxation conditions corresponding to the remaining three sets of McCormick envelope constraints, auxiliary binary variables are introduced in the same way, and two corresponding linear inequalities are set. All the introduced auxiliary binary variables together constitute the set Abv.
[0089] By combining the outer link scheduling constraints, the linearized inner original feasible conditions, the McCormick envelope constraints, the KKT stationary point conditions, the dual feasible conditions, and the complementary relaxation conditions processed by the large constant linearization method, a single-layer mixed-integer linear programming model is obtained. .
[0090] in, This means that when x is satisfied i,m,k Given the feasible values of y, Dual, and Abv, find the maximum value of the expression within the parentheses.
[0091] The single-layer mixed integer linear programming model also satisfies the constraints on the number of laser communication terminals at the satellite end and the ground station end, the constraints on the values of binary link scheduling variables, the two types of robust budget constraints after linearization, the constraints on the physical values of the link state, the boundary constraints of the deviation, the non-negativity constraints of positive and negative deviation variables, the four sets of McCormick envelope constraints, the stationary point condition, the non-negativity constraints of all dual variables, and all complementary relaxation conditions processed by the large constant linearization method.
[0092] Step 107: Solve the single-layer mixed integer linear programming model and output the satellite-to-ground laser link scheduling results.
[0093] In this embodiment, based on the obtained link scheduling variables, the system outputs a list of link establishment targets for both the satellite and ground station in each scheduling time slot, the system throughput under the worst-case link state, and the multi-source deviation corresponding to the optimal solution. The ground control center generates link establishment, maintenance, and release commands in time slot order and sends them to the satellite and ground station before the start of the corresponding time slot.
[0094] For example, a single-level mixed-integer linear programming model is input into a mixed-integer linear programming solver to obtain the values of the binary chain-building variables. The coefficients, constraint coefficients, variable types, and upper and lower bounds of the single-level mixed-integer linear programming model are input into a commercial mixed-integer programming solver, with the solution objective set to maximize the system throughput in the worst-case scenario, and a relative optimal gap threshold ε is set. Preferably, ε ≤ 10. -4 The solver returns the optimal or near-optimal link scheduling variables that satisfy the relatively optimal gap threshold. This embodiment does not limit the specific branching, pruning, or relaxation algorithms used internally by commercial mixed-integer programming solvers.
[0095] The link establishment target for each scheduling time slot is determined based on the value of the binary link establishment variable. When the binary link establishment variable corresponding to the candidate satellite-to-ground laser link composed of the m-th satellite and the k-th ground station in the i-th scheduling time slot is 1, the candidate satellite-to-ground laser link between the m-th satellite and the k-th ground station is taken as the link establishment target for the i-th scheduling time slot. When the binary link establishment variable is 0, the corresponding candidate satellite-to-ground laser link is not taken as the link establishment target.
[0096] Output the list of link establishment objects for each scheduling time slot in the order of scheduling time slots, as the result of satellite-to-ground laser link scheduling.
[0097] For each time slot, the number of links to the same satellite checked does not exceed P, and the number of satellites accessed by the same ground station does not exceed Q. Link establishment, maintenance, and release commands are generated based on the time slot start time, the link to be established, the effective visibility duration, and the terminal number.
[0098] During the execution of the scheduling plan, the actual link availability status and actual transmission rate reported by the ground station or satellite are used to update the nominal values, maximum permissible deviation range and robust control parameters for the next rolling scheduling cycle, thereby forming a closed loop of multi-source state prediction, robust scheduling solution and link building plan execution.
[0099] This embodiment further provides specific examples: The number of satellites M=30, the maximum number of concurrent links per satellite and per ground station satisfies P=Q=1, and the maximum allowable deviation of link availability is a. i,m,k =0.1, maximum permissible deviation of data transmission rate r i,m,k =2 megabits per second, the number of ground stations increased from 5 to 30. A comparison is made between existing deterministic scheduling methods, scheduling methods considering only single-source uncertainty of link availability, and the multi-source uncertainty scheduling method of this application. Simulation results are as follows: Figure 2 As shown.
[0100] Figure 2 This is a comparison chart of system throughput under different scheduling methods. Figure 2 It is evident that the method described in this application achieves higher system throughput than existing deterministic methods and methods considering only single-source uncertainties, regardless of the total number of ground stations. This result demonstrates that as the number of candidate links increases, prediction errors in link availability and data transmission rate accumulate, leading to more erroneous scheduling; simultaneously considering both types of uncertainties can reduce throughput loss caused by multi-source prediction bias.
[0101] In summary, this embodiment provides a satellite-to-ground laser link scheduling method. By simultaneously incorporating link availability deviation and data transmission rate deviation into the scheduling model, two types of budget uncertainty sets are constructed to characterize the feasible range of multi-source prediction deviation. A two-layer robust model is then constructed, with the outer layer selecting link establishment variables and the inner layer selecting the worst-case deviation combination. The two-layer model is transformed into a single-layer mixed-integer linear programming model for solution using deviation variable decomposition, McCormick envelope linearization, KKT conditions, and the large constant method. This ensures the robustness of the scheduling scheme to link state fluctuations while allowing flexible adjustment of the conservatism level through dual robust control parameters. The final output is a scheduling result that can directly guide link establishment and release, effectively improving the throughput guarantee capability and resource utilization efficiency of satellite-to-ground laser communication networks under link prediction deviation scenarios.
[0102] Example 2 Based on the same inventive concept as Embodiment 1, this embodiment provides a satellite-to-ground laser link scheduling system. Since the principle of this system in solving the problem is similar to that of the aforementioned satellite-to-ground laser link scheduling method, the implementation of this system can refer to the implementation of the satellite-to-ground laser link scheduling method.
[0103] like Figure 3 As shown, the satellite-to-ground laser link scheduling system includes: The acquisition module 10 is used to acquire the effective visibility duration, nominal link availability, nominal data transmission rate, maximum allowable deviation of link availability, maximum allowable deviation of data transmission rate, first robust control parameter, second robust control parameter, upper limit of the number of satellite-to-ground laser links that each satellite can establish simultaneously in any time slot, and upper limit of the number of satellite-to-ground laser links that each ground station can access simultaneously in any time slot in the satellite-to-ground laser communication network.
[0104] The first construction module 20 is used to construct the system throughput and scheduling constraints represented by binary link establishment variables. The system throughput is determined based on the effective visibility duration, actual link availability, and actual data transmission rate. The scheduling constraints include the upper limit constraint on the number of satellite-to-ground laser links that each satellite can establish simultaneously in any time slot, the upper limit constraint on the number of satellite-to-ground laser links that each ground station can access simultaneously in any time slot, and the value constraints of the binary link establishment variables.
[0105] The second construction module 30 is used to express the actual link availability and the actual data transmission rate as the sum of the nominal link availability and the deviation of the link availability, and the sum of the nominal data transmission rate and the deviation of the data transmission rate, respectively. It also constructs a set of uncertainties in the budget of link availability and a set of uncertainties in the budget of data transmission rate based on the nominal link availability, the nominal data transmission rate, the maximum allowable deviation of link availability, the maximum allowable deviation of data transmission rate, the first robust control parameter, and the second robust control parameter.
[0106] The third building module 40 is used to construct a two-layer robust model with outer layer selection binary link establishment variables and inner layer selection link availability deviation and data transmission rate deviation, based on system throughput, link availability budget uncertainty set and data transmission rate budget uncertainty set.
[0107] Linearization module 50 is used to address the inner-layer optimization problem in the two-layer robust model. It decomposes the link availability deviation and data transmission rate deviation into positive and negative deviation variables, respectively, to linearize the absolute value constraints. It introduces bilinear auxiliary variables and uses the McCormick envelope to linearize the product of the link availability deviation and the data transmission rate deviation.
[0108] The fourth construction module 60 is used to construct the Lagrangian function and KKT conditions for the linearized inner-layer optimization problem, linearize the complementary relaxation conditions using the large constant method, and merge the linearized inner-layer conditions and scheduling constraints into a single-layer mixed-integer linear programming model.
[0109] The scheduling module 70 is used to solve a single-layer mixed integer linear programming model and output the scheduling results of the satellite-to-ground laser link.
[0110] For more detailed information on the working process of each of the above modules, please refer to the relevant content disclosed in Example 1, which will not be repeated here.
[0111] Example 3 This embodiment provides a computer device, including a processor and a memory; wherein, when the processor executes a computer program stored in the memory, it implements the steps of the satellite-to-ground laser link scheduling method described in Embodiment 1.
[0112] For a more detailed explanation of the above method, please refer to the relevant content disclosed in Example 1, which will not be repeated here.
[0113] Example 4 This embodiment provides a computer-readable storage medium for storing a computer program; when the computer program is executed by a processor, it implements the steps of the satellite-to-ground laser link scheduling method described in Embodiment 1.
[0114] For a more detailed explanation of the above method, please refer to the relevant content disclosed in Example 1, which will not be repeated here.
[0115] Example 5 This embodiment provides a computer program product, including computer-executable instructions or a computer program. When the computer-executable instructions or the computer program are executed by a processor, they implement the steps of the satellite-to-ground laser link scheduling method described in Embodiment 1.
[0116] For a more detailed explanation of the above method, please refer to the relevant content disclosed in Example 1, which will not be repeated here.
[0117] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems, devices, storage media, and computer program products disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple, and relevant parts can be referred to the method section.
[0118] Those skilled in the art will clearly understand that the techniques in the embodiments of the present invention can be implemented using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of the present invention.
[0119] In some embodiments, computer-executable instructions may take the form of programs, software, software modules, scripts, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as stand-alone programs or as modules, components, subroutines, or other units suitable for use in a computing environment.
[0120] As an example, computer-executable instructions may, but do not necessarily, correspond to files in a file system. They may be stored as part of a file that holds other programs or data, for example, in one or more scripts in a Hyper Text Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple co-located files (e.g., a file that stores one or more modules, subroutines, or code sections).
[0121] As an example, computer-executable instructions can be deployed to execute on a single electronic device, or on multiple electronic devices located at one location, or on multiple electronic devices distributed across multiple locations and interconnected via a communication network.
[0122] The present application has been described in detail above with reference to specific embodiments and exemplary examples; however, these descriptions should not be construed as limiting the present application. Those skilled in the art will understand that various equivalent substitutions, modifications, or improvements can be made to the technical solutions and implementation methods of the present application without departing from the spirit and scope of the present application, and all such modifications and improvements fall within the scope of the present application. The scope of protection of the present application is determined by the appended claims.
Claims
1. A satellite-to-ground laser link scheduling method, characterized in that, include: The effective visibility duration, nominal link availability, nominal data transmission rate, maximum allowable deviation of link availability, maximum allowable deviation of data transmission rate, first robust control parameter, second robust control parameter, upper limit of the number of satellite-to-ground laser links that each satellite can establish simultaneously in any time slot, and upper limit of the number of satellite-to-ground laser links that each ground station can access simultaneously in any time slot are obtained in the satellite-to-ground laser communication network. Construct system throughput and scheduling constraints represented by binary chained variables; The system throughput is determined based on the effective visibility duration, actual link availability, and actual data transmission rate; scheduling constraints include the upper limit constraint on the number of satellite-to-ground laser links that each satellite can establish simultaneously in any time slot, the upper limit constraint on the number of satellite-to-ground laser links that each ground station can access simultaneously in any time slot, and the constraint on the value of the binary link establishment variable. The actual link availability and actual data transmission rate are expressed as the sum of nominal link availability and the deviation of link availability, and the sum of nominal data transmission rate and the deviation of data transmission rate, respectively. Based on nominal link availability, nominal data transmission rate, maximum allowable deviation of link availability, maximum allowable deviation of data transmission rate, first robust control parameter and second robust control parameter, a set of uncertain budgets for link availability and data transmission rate is constructed. A two-layer robust model is constructed using system throughput, link availability budget uncertainty set, and data transmission rate budget uncertainty set as the outer layer of binary link establishment variables and the inner layer of link availability deviation and data transmission rate deviation. To address the inner-layer optimization problem in the two-layer robust model, the link availability deviation and data transmission rate deviation are decomposed into positive and negative deviation variables respectively to linearize the absolute value constraint. Bilinear auxiliary variables are introduced and the product of link availability deviation and data transmission rate deviation is linearized using the McCormick envelope. Lagrangian functions and KKT conditions are constructed for the linearized inner-layer optimization problem. The complementary relaxation conditions are linearized using the large constant method. The linearized inner-layer conditions and scheduling constraints are then merged into a single-layer mixed-integer linear programming model. Solve the single-layer mixed integer linear programming model and output the satellite-to-ground laser link scheduling results.
2. The satellite-to-ground laser link scheduling method according to claim 1, characterized in that, Construct system throughput and scheduling constraints represented by binary linked variables, including: Construct the expression for system throughput and scheduling constraints: ; Where ST is the system throughput; I is the total number of scheduling time slots; M is the total number of satellites in the space-to-ground laser communication network; K is the total number of ground stations in the space-to-ground laser communication network; Δt i,m,k φ represents the effective visible duration of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; i=1,2,…,I; m=1,2,…,M; k=1,2,…,K; φ i,m,k η represents the actual data transmission rate of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; i,m,k x represents the actual link availability of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; i,m,k x represents the on / off status of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; i,m,k =1 indicates that the link is connected; x i,m,k =0 indicates that the link is disconnected; This means that the number of links established simultaneously by each satellite in any time slot does not exceed the number P; This indicates that the number of links that each ground station can simultaneously access in any time slot does not exceed the number Q.
3. The satellite-to-ground laser link scheduling method according to claim 1, characterized in that, The actual link availability and actual data transmission rate are expressed as the sum of the nominal link availability and the deviation from the nominal link availability, and the sum of the nominal data transmission rate and the deviation from the nominal data transmission rate, respectively. Based on the nominal link availability, nominal data transmission rate, maximum permissible deviation of link availability, maximum permissible deviation of data transmission rate, first robust control parameter, and second robust control parameter, a set of uncertainties for the link availability budget and a set of uncertainties for the data transmission rate budget are constructed, including: Based on the maximum permissible deviation 'a' of the link availability of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot. i,m,k Construct a link availability budget uncertainty set U1 with the first robust control parameter Γ1; where the link availability budget uncertainty set U1 contains |δ i,m,k |≤a i,m,k , 0≤η i,m,k ≤1, and ;δ i,m,k η represents the link availability deviation of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; i,m,k The actual link availability of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; I is the total number of scheduling time slots; M is the total number of satellites in the satellite-to-ground laser communication network; K is the total number of ground stations in the satellite-to-ground laser communication network; i=1,2,…,I; m=1,2,…,M; k=1,2,…,K; |·| represents taking the absolute value; Based on the maximum permissible deviation r of the data transmission rate of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot. i,m,k The second robust control parameter Γ2 is used to construct the data transmission rate budget uncertainty set U2; where the data transmission rate budget uncertainty set U2 includes: |ζ i,m,k |≤r i,m,k , φ i,m,k ≥0, and ;ζ i,m,k φ represents the signed data transmission rate deviation of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; i,m,k The actual data transmission rate of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot is denoted as .
4. The satellite-to-ground laser link scheduling method according to any one of claims 1-3, characterized in that, A two-layer robust model is constructed using system throughput, link availability budget uncertainty set, and data transmission rate budget uncertainty set. The outer layer consists of binary link establishment variables for selection, and the inner layer consists of link availability deviation and data transmission rate deviation. The model includes: The expression for constructing the two-layer robust model f is: ; Where, x i,m,k x represents the on / off status of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; i,m,k =1 indicates that the link is connected; x i,m,k =0 indicates that the link is disconnected; This means that the number of links established simultaneously by each satellite in any time slot does not exceed the number P; This indicates that the number of links simultaneously accessed by each ground station in any time slot does not exceed the number Q; δ i,m,k ζ represents the link availability deviation of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; i,m,k Δt represents the signed data transmission rate deviation of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot. i,m,k The effective visibility duration of the candidate satellite-to-ground laser link consisting of the m-th satellite and the k-th ground station within the i-th scheduling time slot; The nominal link availability is the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot. Let be the nominal data transmission rate of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; i = 1, 2, ..., I; m = 1, 2, ..., M; k = 1, 2, ..., K; the link availability budget uncertainty set U1 contains |δ i,m,k |≤a i,m,k , 0≤η i,m,k ≤1, and Γ1 is the first robust control parameter; a i,m,k Let |ζ| represent the maximum permissible deviation in the link availability of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; the data transmission rate budget uncertainty set U2 includes: |ζ| i,m,k |≤r i,m,k , φ i,m,k ≥0, and ;r i,m,k Γ2 represents the maximum permissible deviation of the data transmission rate of the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot; Γ2 is the second robust control parameter. This means that when δ is satisfied i,m,k and ζ i,m,k Find the minimum value of the expression within the parentheses among all possible values; This means that when x is satisfied i,m,k Find the maximum value of the expression within the parentheses from all possible values.
5. The satellite-to-ground laser link scheduling method according to claim 4, characterized in that, To address the inner-layer optimization problem in the two-layer robust model, the link availability deviation and data transmission rate deviation are decomposed into positive and negative deviation variables, respectively, to linearize the absolute value constraints. Bilinear auxiliary variables are introduced, and the product of the link availability deviation and data transmission rate deviation is linearized using the McCormick envelope, including: For each candidate link, a non-negative link availability positive deviation variable is introduced. and negative deviation variables and non-negative data transmission rate positive deviation variable and negative deviation variables ;in, ; ; ; ; Using the McCormick envelope, according to The range of values is [-a] i,m,k ,a i,m,k ]and The range of values is [-r] i,m,k ,r i,m,k Constructing linear constraints: ; in, ;ψ i,m,k Let be the bilinear auxiliary variable corresponding to the candidate satellite-to-ground laser link formed by the m-th satellite and the k-th ground station within the i-th scheduling time slot.
6. The satellite-to-ground laser link scheduling method according to claim 5, characterized in that, Lagrangian functions and KKT conditions are constructed for the linearized inner-layer optimization problem. The complementary relaxation conditions are linearized using the large constant method. The linearized inner-layer conditions and scheduling constraints are then merged into a single-layer mixed-integer linear programming model, including: The linearized inner-layer optimization problem is expressed as follows: the link availability positive and negative deviation variables, the data transmission rate positive and negative deviation variables, and the bilinear auxiliary variable are used as the inner-layer original variables. The goal is to minimize the system throughput, and the conditions are met by the deviation boundary constraints, two types of robust budget constraints, actual link availability value constraints, actual data transmission rate non-negativity constraints, and McCormick envelope constraints. For each constraint of the inner-level optimization problem, a non-negative dual variable is introduced to construct a Lagrangian function; the Lagrangian function consists of the sum of the objective function of the inner-level optimization problem and the summation of each constraint multiplied by its corresponding dual variable. The KKT conditions are constructed based on the Lagrangian function. The KKT conditions include: primal feasibility conditions, which are that all constraints of the inner optimization problem are satisfied; dual feasibility conditions, which are that each dual variable is non-negative; stationary point conditions, which are that the partial derivatives of the Lagrangian function with respect to the inner primal variables are set to zero; and complementary relaxation conditions, which are that the product of each dual variable and the corresponding constraint relaxation is zero, and the constraint relaxation is expressed by the inner primal variables. For each complementary relaxation condition, an auxiliary binary variable is introduced, and the large constant method is used to linearize the complementary relaxation condition into two linear inequalities; the large constant is a positive constant greater than the possible values of the dual variable and the constraint relaxation amount. By merging scheduling constraints, primal feasible conditions, dual feasible conditions, stationary conditions, and linearized complementary relaxation conditions, a single-level mixed-integer linear programming model is formed. The single-level mixed-integer linear programming model uses binary link-building variables, inner-level primal variables, dual variables, and auxiliary binary variables as decision variables, with the objective of maximizing the system throughput under the worst-case link state. The worst-case link state is the value of link availability deviation and data transmission rate deviation that minimizes the system throughput within the two sets of budget uncertainty.
7. The satellite-to-ground laser link scheduling method according to claim 6, characterized in that, Solve the single-layer mixed-integer linear programming model and output the satellite-to-ground laser link scheduling results, including: The single-layer mixed-integer linear programming model is input into the mixed-integer linear programming solver for solving, and the values of the binary chain-building variables are obtained. The link establishment target for each scheduling time slot is determined based on the value of the binary link establishment variable. When the binary link establishment variable corresponding to the candidate satellite-to-ground laser link composed of the m-th satellite and the k-th ground station in the i-th scheduling time slot is 1, the candidate satellite-to-ground laser link between the m-th satellite and the k-th ground station is taken as the link establishment target for the i-th scheduling time slot. When the binary link establishment variable is 0, the corresponding candidate satellite-to-ground laser link is not taken as the link establishment target. Output the list of link establishment objects for each scheduling time slot in the order of scheduling time slots, as the result of satellite-to-ground laser link scheduling.
8. A satellite-to-ground laser link scheduling system, characterized in that, include: The acquisition module is used to acquire the effective visibility duration, nominal link availability, nominal data transmission rate, maximum allowable deviation of link availability, maximum allowable deviation of data transmission rate, first robust control parameter, second robust control parameter, upper limit of the number of satellite-to-ground laser links that each satellite can establish simultaneously in any time slot, and upper limit of the number of satellite-to-ground laser links that each ground station can access simultaneously in any time slot in the satellite-to-ground laser communication network. The first construction module is used to construct the system throughput and scheduling constraints represented by binary link establishment variables. The system throughput is determined based on the effective visibility duration, actual link availability, and actual data transmission rate. The scheduling constraints include the upper limit constraint on the number of satellite-to-ground laser links that each satellite can establish simultaneously in any time slot, the upper limit constraint on the number of satellite-to-ground laser links that each ground station can access simultaneously in any time slot, and the value constraints of the binary link establishment variables. The second construction module is used to express the actual link availability and the actual data transmission rate as the sum of the nominal link availability and the deviation of the link availability, and the sum of the nominal data transmission rate and the deviation of the data transmission rate, respectively. Based on the nominal link availability, the nominal data transmission rate, the maximum allowable deviation of the link availability, the maximum allowable deviation of the data transmission rate, the first robust control parameter, and the second robust control parameter, the link availability budget uncertainty set and the data transmission rate budget uncertainty set are constructed. The third building module is used to construct a two-layer robust model with outer layer selection binary link establishment variables and inner layer selection link availability deviation and data transmission rate deviation, based on system throughput, link availability budget uncertainty set and data transmission rate budget uncertainty set; The linearization module is used for inner-layer optimization problems in a two-layer robust model. It decomposes the link availability deviation and data transmission rate deviation into positive and negative deviation variables respectively to linearize the absolute value constraints. It introduces bilinear auxiliary variables and uses the McCormick envelope to linearize the product of the link availability deviation and the data transmission rate deviation. The fourth module is used to construct the Lagrangian function and KKT conditions for the linearized inner-layer optimization problem, linearize the complementary relaxation conditions using the large constant method, and merge the linearized inner-layer conditions and scheduling constraints into a single-layer mixed-integer linear programming model. The scheduling module is used to solve a single-layer mixed integer linear programming model and output the scheduling results of the satellite-to-ground laser link.
9. A computer device, characterized in that, It includes a processor and a memory; wherein, when the processor executes the computer program stored in the memory, it implements the steps of the satellite-to-ground laser link scheduling method according to any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, Used to store computer programs; when the computer programs are executed by a processor, they implement the steps of the satellite-to-ground laser link scheduling method according to any one of claims 1-7.
